whisper_small_tw11

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0462
• Wer: 1.3691

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 6.25e-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1200
• training_steps: 2400

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.5528	2.86	400	2.8688	2.2350
1.5461	5.71	800	2.2548	2.1533
0.6586	8.57	1200	2.4110	1.5250
0.1633	11.43	1600	2.6985	1.4415
0.0318	14.29	2000	2.9465	1.2165
0.0119	17.14	2400	3.0462	1.3691

Framework versions

• Transformers 4.34.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0